What is Merit Pay For?
(cc photo by kevindooley)
Dana Goldstein offers some skepticism about so-called “merit pay” for teachers:
Consider this TED talk on career motivation from Dan Pink, a former Al Gore speechwriter who is now a business journalist. If you can get past the MBA lingo, there’s a lot here that is really consequential for education policy. Forty years of psychological research demonstrates that when someone is faced with a complex, creative task — like teaching — money is an ineffective motivational tool, and may even delay progress. Professionals engaged in creative work are more likely to be motivated by autonomy, and by the feeling that they are part of a larger, socially important enterprise.
That seems plausible to me. But I think it mistakes the purpose of offering higher salaries to more effective teachers. I don’t think the idea is that ineffective teachers are going to suddenly will themselves into becoming great teachers in order to grab some incentive pay. The point is that if you’re employing a bunch of teachers, any of whom might depart in favor of employment elsewhere, you want to make sure that it’s your most effective teachers who are least likely to quit. And one way to do that is to make sure that it’s your most effective teachers—rather than simply your longest-serving ones—who are getting paid the most money.
Indeed, for all the controversy around differential pay schemes at some level I don’t think even the most old-school of teacher’s union leaders seriously dispute this logic. After all, it’s extremely common for collective bargaining agreements to offer enhanced salaries to teachers who have more educational credentials. The logic here, presumably, is that more educated teachers are more effective teachers and thus it makes sense to pay extra to retain them. The diplomas, in other words, are a proxy for quality. Similarly, veteran teachers get paid more than brand new teachers on the theory that a more experienced instructor is a better instructor. The principle that it makes sense to pay extra for quality isn’t seriously in dispute. The problem is that diplomas and time served turn out to be bad proxies for quality: “Recent research, however, suggests that such paper qualifications have little predictive power in identifying effective teachers.”
The reform proposal, ultimately, isn’t all that radical. Rather than paying extra for very weak correlates of effective teaching, why not just pay extra for effective teaching? To the extent that such a compensation scheme creates incentives for teachers to improve their own performance, that will be nice. But the real benefit to paying for quality is that, over time, it will encourage effective teachers to keep teaching while encouraging ineffective teachers to find jobs to which they’re better-suited, thus improving the overall quality of the instructor pool.
68 Responses to “What is Merit Pay For?”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
September 8th, 2009 at 4:06 pm
The usual faux-naive education reform rhetoric is slightly annoying. The reason it’s not so simple as “let’s just pay according to a better proxy” is that the candidate for better proxy might just as well turn out to be “willing to teach science students that the world was created in six days.” That’s what happens when Sarah Palin gets on the school board.
September 8th, 2009 at 4:07 pm
Matt, I know plenty of other people have said this, but…
Kindly, make more of an effort to be actually knowledgable about education.
I would much rather prefer a book review from you about Lev Grossman’s The Magicians. It seems really up your alley.
September 8th, 2009 at 4:10 pm
Let me be the millionth person to point out that we can’t get agreement on what successful teaching is. Given this, giving some people more money than others on seemingly arbitrary grounds creates a most unhealthy work environment.
September 8th, 2009 at 4:12 pm
I’ve been teaching for 20 years at a community college and the only thing I can say about this debate is that unless you come up with an objective benchmark to compare one instructor to another across academic disciplines you’ll never find a solution to the problem. I’ve been around Ph.D’s that can’t teach their way out of a paper bag and around freshly minted Masters instructors who don’t know the first thing about teaching. All I can say is that some people have it and some people don’t. And that’s not objective at all.
September 8th, 2009 at 4:13 pm
Most people who teach went into teaching because they wanted to be teachers. There are better paying professions, but teachers get three months off a year and for certain personality types it’s the only job they could imagine having. The idea that, our most skilled teachers will go work for NASA or something is far fetched. Higher pay would go to those with seniority or folks more skilled at teacher politics. It’s basically impossible to quantify ‘effective teaching’ – much like ‘effective policing’ or ‘effective governing’ – its something you know when you see it but can’t really establish criteria for identifying it. All ‘merit pay’ will lead to is further strip-mining of taxpayers and no noticeable improvement in student performance. Just say no to merit pay.
September 8th, 2009 at 4:15 pm
Merit pay is a way for “reformers” to try to fellate teacher-bashing conservatives. I would have thought that was obvious.
September 8th, 2009 at 4:16 pm
As usual Matt waves his hands and talks about “teacher quality,” as if we know what that is, and grumbles about those dastardly teachers’ unions who won’t let teacher pay be based on quality.
If unions (where they are powerful, which is certainly not everywhere) want teachers to be paid more for seniority and educational attainment, why are they so doggone unreasonable about the obviously much better concept of “quality.” Are they just big stupid jerks who don’t care about the children?
Or is it possible that “quality” is not something reasonably measurable and is also quite easily games by school administrators to help them wage their petty turf wars and political disputes? Is that possible?
Teacher “quality” is almost immaterial to student performance–so long as the teacher is a functional and literate adult. Student “quality” (by which I mean educational attainment and economic status of the student’s parents) is far more important. What else explains the fact that highly selective private schools pay their teachers far less than public schools? Do they know something they didn’t allow MY to be taught?
September 8th, 2009 at 4:18 pm
There is good data correlating teacher IQ with student performance. Not so with any sort of teacher education or credentialing.
I do agree that pay will not likely affect the performance of a teacher — their relationships with students have a much stronger effect (at least on me, albeit I teach grad school level, but I bet it is the same for k-12).
What pay might influence is which people choose to go into teaching. Right now our elite students go into Ibanking or law school or medecine. Higher pay might make some of them reconsider. Worth testing, anyway.
September 8th, 2009 at 4:20 pm
Since the purpose of teaching is to improve standardized test scores, there is no difficulty in determining who the best teachers are.
September 8th, 2009 at 4:20 pm
Because we largely don’t know what it is, except in a tautological, Justice-Potter-Stewart-and-pornography way. Or if you define ‘effective’ as ’score-increasing’. At which point it’s easy.
Everyone in the debate is resorting to weak correlates.
And besides, didn’t we figure out years ago it’s the damned teachers’ unions?
September 8th, 2009 at 4:22 pm
The Lorax comes very close to my objection — but it’s worse than even that. Not only can we not agree on what effective teaching is there really is no way to accurately assess it. My mother has worked in the Fairfax County Public Schools and Montgomery County Public Schools. She teaches special education. Most of her students will not “succeed” by the metrics we would normally use to assess student performance. Is she a poor teacher because of that?
How about this: I have no degree. I have some informal training in computers and a minor certification. My salary is low for my field and I still make more than most entry-level teachers. With more training and experience I can make more than most teachers ever will.
I knew that growing up because my folks were both teachers. Most of my friends knew that growing up because our teachers complained and television made fun of teachers not making much money. As a result most of us chose to get into careers as far away from teaching as possible. I even had some interest in teaching, but teaching seemed like a mission rather than a career. Couple that with the licensing requirements and the social stigma attached to teaching (”those who can, do. those who can’t, teach” is really a nasty thing to say about folks who socialize your kids) and you’ve got some fairly hefty disincentives to teaching.
I think if you did more to reduce those barriers and made teaching more accessible (and lucrative) for folks, you’d have more interest. If I wanted to go into teaching at this point I’m pretty much going to have to give up working for 6 months (student teaching–while I might be able to get part time work with a school system I will not be able to work full time) and then compete for a new job. After I win that job I will be paid less than I make currently for comparable work with an expectation I will work from home most nights.
For that I get 2 months unpaid leave in the summer. I’m surprised we have trouble finding teachers.
September 8th, 2009 at 4:31 pm
Even if you take the simplest example, grade school teachers, it seems impossible to me to have a fair merit pay system. Do you judge them by the results of testing? What if one teacher has a reputation for working well with troubled students and therefore gets a higher percentage of them? Then you are taking money from that teacher at the same time you are making the job much more difficult. And don’t even think about administrators or school boards figuring out which teacher is best.
September 8th, 2009 at 4:35 pm
[I'm getting really annoyed at this comment system. It has been 25 minutes since I posted my last comment, so why does it think I am "posting too quickly"? And why does it erase my text instead of just taking me back to the edit screen?]
OK, my long comment got erased. Wall Street gave itself 11 years, from 1998-2009, to evaluate the effectiveness of its “merit pay” system and its “innovative new ideas”. And on Wall Street an incentive payment of $400,000/year is considered an insult worth quitting over. So let’s make teacher base pay be $175,000/year and put in place an incentive payment of up to $275,000 for a total of $400,000/year. We’ll test that for 11 years and see how it works.
Cranky
September 8th, 2009 at 4:36 pm
The point is moist merit pay isn’t “good job! Here’s a bonus!” Its like the DC plan you were so gung ho for “We’re taking away your seniority and adding in the ability to fire you for political reasons if need be. In addition we’re cutting your base pay. But if you do a good job you might be able to earn that back with a bonus! Well until we run out of the grant we’re using for bonuses anyway.”
September 8th, 2009 at 4:37 pm
In my childhood in Oklahoma, which was a long time ago, there was merit pay for teachers. In theory. In practice, it went to the football coaches. The local education association opposed it, but not exactly out of greed. And if not a winning record at football, how to judge? Leave it up to administrators? Give me a break. I thought I knew who my best teachers were and I doubt that anyone agreed.
September 8th, 2009 at 4:47 pm
This isn’t really my issue, but I have to say, all the handwaving about how we can’t possibly assess teacher quality strikes me as pretty unconvincing. You know what? It’s hard to assess quality among most white collar jobs. For any employee who doesn’t work in sales, evaluating performance is kind of a bitch. Yet just about every company in the world does performance reviews, because, despite their huge and obvious flaws, performance reviews on balance are useful.
Nothing about teaching strikes me as so categorically different than every other job in the world.
September 8th, 2009 at 4:49 pm
Let me chime in here with the other standard objection since so many people have already pointed out (yet again) the difficulty of assessing merit:
The problem is not only what is considered meritorious, but who is deciding this and handing out the goodies accordingly. If you get a proposal for merit pay that puts the authority in the hands of teachers and boards controlled by teachers, that ’s one thing. But every faux-concern proposal I’ve seen puts the power of assessment and the power of the purse in the hands of the administrators and the school boards. Does anyone seriously doubt that these new-found powers will be used as a club in the hands of the latter parties? And that the club will be weilded to enforce various political agendas?
If Matt claims not to have considered this, then he really shouldn’t be opining on the subject at all now, should he?
September 8th, 2009 at 4:52 pm
The only possible system I see having any possibility of fairness would be some sort of evaluation by the entire staff of a building for each staff member. No one knows if you raise or lower the performance of your school better than your colleagues.
You can talk performance deltas, staff and student attendence, and all sorts of other criteria, but if your colleagues collectively think you are a benefit to the school, you probably are, and vice versa.
But generally I think the whole idea is an exercise in futility. I’ve yet to see a merit pay structure or proposal that seems fair. And most of the dollar amounts I’ve seen have been less than enticing.
September 8th, 2009 at 4:53 pm
But now, it doesn’t matter how good that teacher is, they won’t get any more money than a crappy teacher who shows filmstrips all day.
Why not? Right now it’s teachers vs. administration. Administration can’t punish or reward good teachers. A shitty school board can use their power to do what? Mess with the menu at the cafeteria? No wonder it’s filled with idiots.
September 8th, 2009 at 4:55 pm
Eh: exactly.
September 8th, 2009 at 4:59 pm
Another benefit of tackling performance based pay is that at least it turns educators’ attention to _data_. Data driven instruction would not be such a bad thing. The first effect it would have is to stimulate demand for better data and better methods for analyzing and summarizing the data. Before long, it might actually translate into a culture of continuous improvement toward measurable goals.
There is no question that education outcomes are difficult to assess — not just student outcomes but the unique contributions that each teacher makes to student outcomes, above and beyond the influences of the students’ home and previous teachers.
But just because they are difficult to assess doesn’t mean we shouldn’t try. Until then, we’ll continue to screw over and discourage the truly outstanding teachers.
September 8th, 2009 at 5:00 pm
I believe “Merit Pay” is a tool that will be used to force good teachers away from poorer “failing” school districts and towards richers “successful” ones.
That’s what it’s for, isn’t it?
September 8th, 2009 at 5:00 pm
@Eh
People getting pay raises for reasons other than merit and meritorious people not getting them is a constant source of problems in the workplace. I don’t know why you’d want to add that to the already shitty conditions that most teachers face today. I mean, there’s all manner of things that occur commonly in the workplace that we don’t want to import into our schools.
I do think that teaching is different than many jobs. You got all your reports in on time and kept your ass off the company copier–that’s something that is pretty easy to assess. You did a good job teaching these children–that’s something much harder to assess.
September 8th, 2009 at 5:00 pm
This:
September 8th, 2009 at 5:01 pm
Or because their expensive management consultants told them so, despite a complete lack of evidence.
I’m always touched by the childlike faith some people have in the rationality of corporate bureaucracies.
September 8th, 2009 at 5:06 pm
I’ll buy into this as soon as someone can tell me what makes a “successful teacher”.
The current theory seems to be that a good teacher is someone who can deliver better standardized test scores. But standardized tests are a terrible metric of anything other than a student’s proficiency at taking a standardized test, and have a deleterious effect on the curriculum and educational system as a whole.
We need to abandon this notion that we can objectively measure something that is, for the most part, impossible to objectively measure.
What’s needed is more teacher training and better qualified teachers overall – people who are subject matter experts, as well as experts in the relevant areas of psychology, as well as excellent presenters of information. You’re not going to get this without substantially raising salaries across the board, and starting to treat teachers like highly qualified highly paid professionals a la a doctor or lawyer.
September 8th, 2009 at 5:09 pm
“You’re not going to get this without substantially raising salaries across the board, and starting to treat teachers like highly qualified highly paid professionals a la a doctor or lawyer.” @Eric
And of course there’s no reason why they can’t be treated as such, just as there’s no reason why we can’t unionize big box store workers and have them paid more. This status quo certainly wasn’t inevitable, though it no doubt is tied to the fact that teaching was seen as women’s work.
September 8th, 2009 at 5:12 pm
I think you overlook an important point when talking about merit pay. When teachers list reasons for leaving the profession the pay for teaching is towards the bottom. Teachers that leave are more upset by administration, overloaded classrooms, and lack of support. I think many teachers would stay if those issues were dealt with. Doing some of them might cost some money but make teaching easier. Another reason teachers leave is because of lack of respect from the rest of the community. Until we change the culture teachers will continued to be devalued and therefore leave for something they feel more valued doing.
September 8th, 2009 at 5:13 pm
My third-grade teacher was a soft-spoken pretty woman with a French surname who allowed the boys to leap out of their seats like leopards when answering multiplication questions. She was having as much fun encouraging the constructive rowdiness as we were being rowdy. Whatever the criteria for merit pay raises are – it should start with someone like her.
September 8th, 2009 at 5:15 pm
Forty years of psychological research demonstrates that when someone is faced with a complex, creative task — like teaching — money is an ineffective motivational tool, and may even delay progress.
Except for the fact that research also shows that the best method of teaching isn’t being complex and creative, it’s Direct Instruction, which is highly scripted–exactly the sort of task that Pink concedes is suited to improvement via incentives.
September 8th, 2009 at 5:25 pm
One thing that should be looked at is how seniority rules are used to allocate assignments. In practice, here’s what happens: the newest teachers with the least experience get the worst (meaning, lowest skilled students) classes, which raises the probability of those low skilled students getting more bad teaching (1st years teachers, like 1st year anything, make the most mistakes).
The biggest obstacle to such changes: the unions.
September 8th, 2009 at 5:26 pm
Sigh. Basic economics – forgone costs. Even assuming the fallacy of bifurcation here, if it’s ‘a crappy teacher who shows filmstrips all day’ vs ‘we are going to take your seniority away from you and pay you according to how well you toe the line politically’, I know which one I’d take.
Here’s a thought: right now teachers can’t punish or reward good administrators[1]. How about we make administrative pay subject to this merit system . . . and let the teachers decide who is meritorious, and who gets the bonus? Sounds good, right? What could possibly go wrong with this type of incentive?
[1]Actually, there’s quite a bit administrators can do to punish teachers already. Starting with teaching assignments and working up to non-renewal of contract. Here’s something I like to link to when this twaddle comes up:
September 8th, 2009 at 5:41 pm
Apparently, poorly designed merit pay systems can have unanticipated consequences:
http://books.google.com/books?id=7gejitbXx3MC&pg=PA144&lpg=PA144&dq=sway+community+high&source=bl&ots=YPs_8aVxn-&sig=VS8BQyC24Z_VuNKX9lVBEIWS01M&hl=en&ei=rM2mSv36O4KgnQeB6KC_Bw&sa=X&oi=book_result&ct=result&resnum=1#v=onepage&q=&f=false
According to the book, anticipation of a potential material reward can crowd out altruistic motivations.
September 8th, 2009 at 5:57 pm
Matt says:
“After all, it’s extremely common for collective bargaining agreements to offer enhanced salaries to teachers who have more educational credentials. The logic here, presumably, is that more educated teachers are more effective teachers and thus it makes sense to pay extra to retain them.”
No, the real logic is more to erect barriers to entry.
September 8th, 2009 at 6:04 pm
These days, everybody is in favor of having Better Teachers in our public schools: Barack Obama, Arne Duncan, Bill Gates, the whole gang. Everybody is in favor of hiring Better Teachers and easing out Worse Teachers.
Heck, I’m in favor of Better Teachers.
But guess what Obama et al haven’t figured out yet about Better Teachers? It’s something that James S. Coleman figured out in working on his 1966 Coleman Report.
An important source is “Race and Education: 1954-2007″ by U. of Delaware historian Raymond Wolters. A major figure in Wolters’ book is quantitative sociologist James S. Coleman, who was given $1 million by the 1964 Civil Rights Act to study how much blacks were shortchanged by the public schools. But his 1966 Coleman report proved disappointing to LBJ Administration. Wolters writes:
” The achievement gap troubled Coleman. As a sociologist he was inclined to ascribe the differences in black and white test scores to the influence of the social environment, and he also knew that attributing even part of the difference to racial inheritance would place him outside the pale of his profession and render him ineligible for future frants. For Coleman and for many other educators and sociologist who studied his report, the key variables were family background and neighborhood. There was no correlation between test scores and per-pupil spending, age of textbooks, and a host of other measures. But there was a correlation with family background, the education and occupations of parents, and the number of books in the home. …
“For Coleman, these findings were unwelcome. Personally, he favored more spending for education. And Coleman’s dismay was compounded by another correlation that emerged from the data. Both black and white children seemed to do better on tests if their teachers had done well on a standard test of vocabulary. This was especially problematical because black teachers were “on the whole less well prepared, less qualified, with lower verbal skills, than their white counterparts.” This led to “the conjecture that [students] would do less well on average under black teachers than under white teachers.” If so, “a major source of inequality of educational opportunity for black students was the fact they were being taught by black teachers.” Yet this possibility was so heterodox that the Coleman report did not pursue the matter. In 1991 Coleman expressed regret over the decision “not to ask the crucial question.” “A dispassionate researcher,” he wrote, “would have gone on to ask the question we did not ask.” …
” Poring over the statistics, he noted that African American teachers, on average, had slightly more years of formal education than their white counterparts. But the black teachers lagged behind whites in vocabulary and reading comprehension.”
In other words, what Obama hasn’t figured out yet, although James S. Coleman figured it out back in 1966, is that, on average, Better Teachers means Whiter Teachers.
When it finally dawns on Obama that if we actually start firing worse teachers and hiring better teachers, we’ll be, on net, firing blacks and hiring whites, you can expect this whole effort to get buried so far under affirmative action that nothing good comes of it.
September 8th, 2009 at 6:07 pm
It’s widely believed that if only we got rid of teachers unions, then we’d have superstar teachers in every inner city classroom. Yet, history suggests that we might wind up with worse teachers because rising politicians would try to fire the old white teachers and give their jobs to co-ethnics.
That’s exactly what happened in the late 1960s in the black Ocean Hill neighborhood in New York City, when the NY school board temporarily decentralized. Black politicians immediately fired huge numbers of white teachers (mostly Jewish) and hired blacks. Albert Shanker, the union boss of the United Federation of Teachers, went on the warpath. A huge brouhaha ensued and Shanker eventually mostly won and got the white teachers re-installed. In “Sleeper” (1973), Woody Allen is told by the people of the future that his age had been obliterated when “a man named Albert Shanker got hold of a nuclear warhead.”
That teachers unions and their seniority rules keep white teachers in jobs in minority-run cities is one of those phenomenon that nobody talks about but is staring you right in the face.
September 8th, 2009 at 6:33 pm
The idea that it is impossible to work out a metric for teacher effectiveness is ridiculous. Plenty of large organizations figure out ways to monitor the effectiveness of their employees. There is no reason to think you couldn’t work up some formula which combined objective measures like test score improvments and subjected measures like supervisor and peer assesment.
September 8th, 2009 at 6:33 pm
This is a lot like the healthcare debate.
There are real-world examples of systems that work better than ours – but they are not even discussed for political reasons. The Japanese have much better results and so do the Germans. What is the difference?
September 8th, 2009 at 6:50 pm
The problem with merit pay is that it pre-supposes that principals and the rest of the administrative heirarchy who actually understand and value the best teachers at their school. This is not necessarily, or even usually, the case. Principals are managers, just like in any other work environment. More than that, they are public employees, who operate in a context where the avoidance of a political or publicity isue is the most important objective. They value loyal employees who do not show them up, make them feel stupid, expose their inadequacies or cause trouble. People in that box, particularly at very bad schools, are always the worst teachers who view their job as a source of a paycheck and nothing more. For someone who watches so much of The Wire, I would think this would be a more intuitive point for you, Matt. The lack of this parallel in the schools in the fourth season is, by the way, one of the only places where the show really pulled a punch and one of my biggest issues with it. To watch the fourth season and accept its view of the problem is to basically assume that it’s all the fault of those damn, drug-addicted negroes and their lack of interest in the education of their children. The schools suck b/c the school administration is ten times as incompetent as any police department. But, for whatever reason, they refused to delve into that.
September 8th, 2009 at 6:51 pm
ron, you can say that again. This country may well die of NIH syndrome. Our insularity and arrogant incuriousness about how things are done in the rest of the world is beginning to remind me of the decline of China in the Ming era.
September 8th, 2009 at 6:56 pm
Matt, I have to believe that if you’ve read this far you understand how badly you blew it. Very surprising departure from your usual well thought out commentary. Why risk your reputation on sloppy work?
September 8th, 2009 at 7:09 pm
9
Since the purpose of teaching is to improve standardized test scores, there is no difficulty in determining who the best teachers are
Actually there are difficulties. The easiest way to raise test scores is to cheat. Presumedly this isn’t what you have in mind (else we can just give everybody a perfect score and go home). So you can’t have teachers testing their own classes. But secure independent testing is expensive.
Assuming this problem is solved you have a random noise problem. Classes are small enough that a teacher may appear to do well or badly just because of random noise.
Another problem is the effects of different teachers fade over time. A third grade class with a good teacher may score better than a third grade class with a bad teacher when tested at the end of third grade but by the time students finish sixth grade the differences will have largely disappeared.
September 8th, 2009 at 7:20 pm
I think that Matt Y. and most comments here have completely missed the point of the article he links to. Re-read the quote at the top, or read the linked article, or best of all go watch the TED video that Goldstein links to.
HOW you measure “teacher quality” is not the issue. The point of Dan Pink’s talk is that EVEN IF you could accurately measure teacher quality, rewarding those teachers with more money would NOT be the right idea. Monetary rewards HURT creativity, according to years of research cited by Dan Pink.
Merit pay–even if you could do it accurately–doesn’t increase fresh thinking. Monetary incentives work when the task at hand has only one correct answer or method–and that certainly does not describe teaching.
September 8th, 2009 at 7:30 pm
What Matt is saying is something like looking at a field hospital in a jungle, where the patients come in with malnutrition, open sores, and long-standing diseases and wounds that won’t heal, to operating rooms and wards that are filthy and understaffed, and thinking that if you pay the doctors who do the most operations more, you will eventually improve the health of the region.
And what it all tells us is that if you want to impress people, make a lot of money, and act important, go to Harvard. If you want a real education, and feel secure enough to afford such a luxury, go somewhere else.
In any case, leave the heavy lifting to parents and teachers who actually do the work, if you don’t have any good ideas to contribute.
September 8th, 2009 at 8:19 pm
> Actually there are difficulties. The easiest
> way to raise test scores is to cheat.
Nah, you don’t have to cheat. You push out just enough of your low-achieving students so that your test scores go up but not so many that your demographics rise to the next band. Instant NCLB bonus!
The best way to manage this is to keep a pool of scholarship students from a nearby failed urban district (Detroit, DC) on hand. This lets you take lots of nice feel-good pictures about how benevolent you are being and also provides a pool to adjust the scores with as needed. I can point you to several large well-regarded suburban districts in the midwest that are playing this game right now.
Cranky
September 8th, 2009 at 8:23 pm
> Matt, I have to believe that if you’ve read this
> far you understand how badly you blew it. Very
> surprising departure from your usual well thought
> out commentary. Why risk your reputation on
> sloppy work?
Yglesias has stated explicitly that he has no experience in teaching K-12 teaching and has no plans to ever volunteer at his local DC elementary school. He also has no experience in actual corporate work (that is, other than as a writer where he was semi-independent) and no experience in corporate management. He is convinced he and Mr.(E) Klein are right about how to “fix” schools and he won’t acknowledge error in these threads because he doesn’t think he has made any. Teaching actual children is not something he needs to dirty his hands with.
Cranky
September 8th, 2009 at 8:49 pm
Interesting. The piece actually seems to advocate paying teachers more at the outset of their careers, rather than back-loading the majority of their pay towards the end.
It points out a problem with a merit pay system for teachers – that it would set them into competition with each other, which would hinder the transfer of skills from more skilled to less skilled teachers, and strain the working relationship of the faculty at the school. In addition, the TED lecture actually says that according to social science research, incentive pay would not work, but would rather decrease performance, for a task like teaching.
I would not have guessed the content of the article, or the content of TED lecture, from the comments or Matt’s response.
Matt seems to think that, despite all these problems, merit pay would cause more effective teachers to stay around. Maybe it would, although one wonders, would there be enough effective teachers who stay to offset those who leave? In a profession that already has shortages? I realize people like to cite, dubious in my opinion, research that suggests that classroom size makes no difference, but I think that, even if those dubious studies were granted to be true, beyond a certain point, under-staffing would begin to stress further schools that are already experiencing difficulties. Such a phenomenon would occur in pretty much every other organization dedicated to task- I fail to see what makes schools different. As I have said before, its the wrong way to the look at the problem.
Instead, after we decide what an effective teacher is, we should ask ourselves- how do we replicate teachers that we deem effective, that is, how do we train and empower someone willing to work hard to become an effective teacher, and, how do we attract people willing to put in that kind of work to the profession?
September 8th, 2009 at 9:13 pm
What a lot of this discussion misses, I think, is the fact that most people are happy with their schools, most students do well, most schools are good, and most teachers do a fine job.
A good school can handle a few bad teachers without seriously compromising education. My 11th grade physics teacher was absolutely nuts and taught us essentially nothing. I also had a “history” teacher and a “biology” teacher who were really coaches with little knowledge of teaching. They imparted some facts and gave some tests but also often conveniently left the room during tests so the students could cheat (mostly off of me).
Not to condone this sort of thing, but this is not what we’re talking about here. No middle class child of college educated parents is seriously harmed because his 9th grade biology teacher was an idiot. In fact, learning to deal with idiots in a position of power is valuable in and of itself.
The real education problem in this country, such as it is, is with poor, mostly minority, mostly urban schools. These represent a relatively small percentage of the public school system. So talk of good vs. bad teachers should not lose sight of what we’re really talking about here.
Steve Salier, amidst his racist insanity, actually touches on an important point. The income level and educational attainment (and stuff like the number of books in the household) are the real determiners of student performance–not teacher “quality”. Switch the entire student body of a New England prep school and a Bronx public school for five years and the results for the students will be about the same. The rich kids whose parents went to Harvard will do well. The poor kids whose parents never graduated from high school will do poorly.
The point is not that there’s nothing to be done. The point is that we’re looking at the problem the wrong way around.
Our public education problem is a poverty problem.
September 8th, 2009 at 9:26 pm
Rob Mac @ #48,
I’ll agree with you up to a point. Having seen the results when a middle-middle suburban school district focuses it best efforts on 2nd, 3rd, and 4th grade voluntary transfer students from an inner city district in an attempt to bring them up to somewhere near grade level, it is clear that the first 2-3 years of childhood are crucial and it is very hard for any child who doesn’t get the background then to recover from that.
But the counterpoint is this: we are generally talking here about 1 teacher per classroom of 10-30 kids, maybe one TA split between 4 or 5 classes, a bit of specialist support. And in Matthew Yglesias’ fondest dreams these so-called merit payments amount to 10-20k, bring that teacher up to a munificent 60, 70, even 80k/year in salary! Whoo hoo!
Meanwhile, I have sat in corporate meetings to discuss fairly minor issues (real issues, not unimportant, but not major either) with 10 people from the company, 5 lawyers, and 5 consultants. A few 10-hour days like that adds up to 80k pretty fast and no one blinks. A merit bonus on Wall Street or LaSalle Street better be at least $5 MILLION or the person receiving it will walk. And so forth.
So what would happen if we put 1 teacher and 2 TAs in EVERY disadvantaged classroom? Paid the teachers 125k BASE and up to 200k merit? What would happen then? But of course we don’t know, because that’s “not realistic”. It is OK to pay a Thomas Friedman 100k _per speech_, but for some reason teachers – esp K-5 teachers – are saints who should be expected to give their heart and soul for 35k/year. Who else is expected to give their heart and soul? Oh yeah, doctors. How many doctors do you know, even at indigent clinics, who settle for $35k/year?
Cranky
September 8th, 2009 at 9:34 pm
The problem is that diplomas and time served turn out to be bad proxies for quality: “Recent research, however, suggests that such paper qualifications have little predictive power in identifying effective teachers.”
So, one think tank pundit quotes another think tank “study” to back up a claim? An epistemological circle-jerk… nice.
If you look at some examples from the real world of peer-reviewed scholarship the picture is far more complicated, and it certainly does not back up this assertion.
Linda Darling-Hammond et al. “Does Teacher Preparation Matter?” Education Policy Analysis 2005
Sharon Kukla-Acevedo “Do teacher characteristics matter?” Economics of Education Review 2008
Ronald Heck “Examining the Relationship Between Teacher Quality as an Organizational Property of Schools and Students’ Achievement and Growth Rates” Educational Administration Quarterly 2007
MY: Quit blogging about education… It’s disgraceful.
September 8th, 2009 at 9:56 pm
@ 48 and 49
Thats a pretty depressing assessment. Although I don’t doubt that there is some truth to what you are saying, in the charter school where I worked, that served a similar demographic (it could not be selective with who it admitted), although there certainly were kids who will always have problems in an educational environment, there were also others who would do well in any school, even some who would do much, much better than average anywhere they went. Maybe even do well at the Dalton School and graduate as a philosophy major from Harvard
In any case, we have to try. We don’t really have any other choice. In my perfect world, i guess school district boundaries could stay the same, but there would have to be national funding of education to equalize per pupil funding between urban, suburban, and rural districts, with additional funding provided to the urban and rural districts to deal with the effects of poverty. Teaching would be a 2 year graduate degree, plus a one year paid internship, with programs required to adhere to strict curricular requirements in order to ensure that they are teaching only practices that are proven effective.
Once in their schools, the teachers would serve a three year apprenticeship, during which time they would be evaluated by a panel composed of other teachers, a member of the administration, and a member of the community. All evaluations would be accomplished in this way. Pay would range from 60-100 thousand dollars (40 thousand for the internship), with some adjustment on either side for cost of living.
Basically, I would turn teaching into a real profession.
September 8th, 2009 at 10:08 pm
I am a huge union supporter and a believer in public education. And I think that those of you self-righteously castigating Matt for his post are full of it.
One commenter writes: “Monetary incentives work when the task at hand has only one correct answer or method–and that certainly does not describe teaching.”
What a blooming idiot!! Do you think that engineering problems have “only one correct answer or method,” or that computer programming questions have “only one correct answer or method”?
They don’t. Instead, like tons of other fields — from science, to finance, to law, to sales — people come up with varied, successful solutions using creativity and hard work. Just like teaching.
You know what? Monetary rewards do work in those fields. At the very least, schools should be making an effort at the margins to make sure that good teachers don’t hop out of education into other fields because of the money. (What about if bad teachers leave? Good riddance! We need bad teachers — like the ones sitting in the NYC rubber rooms profiled in the recent New Yorker article — to go do something else instead of bleeding our local governments dry and screwing up millions’ of childrens’ education.
At the very least, isn’t it worth experimenting with merit pay in some jurisdictions? The unwillingness of this blog’s readership to even consider changing things up, when the horrendous failings of public education as is are obvious for all to see, speaks volumes about some people’s ideological blinders.
If you can’t see that public education, as designed by unions and ed schools, is broken, then you are simply impervious to empirical testing of your ideological suppositions. I can understand why the teachers’ unions refuse to change things up — the system works fine for them. But I can’t see why others are willing to sacrifice the education of millions of children just to shore up their failed ideological dreamworlds.
September 8th, 2009 at 10:09 pm
” there would have to be national funding of education to equalize per pupil funding between urban, suburban, and rural districts,”
And you think those who have are screaming now about health care? Just try equalizing school funding. Soccer moms would be (wo)manning the guillotine.
It’s really fucking pathetic and wrong. I need to get out of Sweden; all I see is what the US could be but never will be because of how ignorant and selfish too many Americans are.
September 8th, 2009 at 10:24 pm
Matt, both you and Dana are right. There seem to be three main reasons that people advocate merit pay:
1.) to encourage teachers to work harder/do a better job (Dana’s argument)
2.) to recruit and retain more successful teachers (your argument)
3.) to create a fairer system in which those who are better get paid better (oops, you guys forgot that one)
Beyond all the problems surrounding the who and how of teacher evaluations, there are a couple things that most people seem to forget or ignore:
1.) The vast majority of value-added measures are not ready for prime time — they’re statistically close to useless. Meaning that ratings based on them are as well. This isn’t to say that they can’t be made better, just that with the current tests and methods, that that’s reality.
2.) Only about 1/3 of teachers teach a tested subject. So even if we devise a precise and foolproof ratings system based on standardized test scores, that leaves 2/3 of the teaching workforce that need to be evaluated in some other way.
September 8th, 2009 at 10:37 pm
@ dcuser
Before you start calling people idiots, and accusing them of being impervious to empirical data, it might serve you well to actually read the article that Matt was responding to, as well as watch the TED talk that it was based on.
I myself am very much concerned with fixing education in this country. I just want the fix to be done right, and not be sacrificed to faddish thinking driven as much by ideologically convenience as by actual science and careful thought. I would not be as certain that the empirical evidence is on your side as you evidently are.
BTW- Merit pay is being tried in places.
September 8th, 2009 at 11:39 pm
50
Here is a paper that says certification makes little difference.
September 9th, 2009 at 12:20 am
@ the truth
Thanks for the support. (I’m the blooming idiot you defended.)
@dcuser
As “the truth” points out, you need to take another look at the article Matt links to. You, Matt, and the majority of commenters here have missed the point entirely.
You don’t have to agree with it, but the research that Goldstein refers to–Dan Pink’s TED lecture on motivation–shows that merit pay and bonuses of the kind offered by the engineering and computer firms you refer to does not work. Many studies, in fact, show that it is counter-productive. Offering people monetary rewards makes them less creative and narrows their thinking.
Supporters of merit pay for teachers need to defend the general proposition that monetary bonuses are a good idea, for any employee of any kind of firm. The research Matt links to–and completely ignores–argues to the contrary.
The reason so many people are missing this is because–as Pink points out–the idea of merit pay has become entrenched conventional wisdom in the business world, despite the fact that there is a large body of evidence that shows it is ineffective.
You may be thinking, “Of course merit pay and bonuses make people do better work,” but, again, the point of the research cited in Matt’s link is that is does not lead to a better work product.
The video is 18 minutes long, and it’s a good lecture. Take a look and see what you think of the argument.
September 9th, 2009 at 12:28 am
[...] 9, 2009 · Leave a Comment Matt Yglesias defends the idea of merit pay for [...]
September 9th, 2009 at 12:36 am
Presumably pointless to chime in at this late point, but I’ll offer this late (and unsupported) claim: teachers don’t actually matter. At least not in a statistically important way. Kids do what their parents require of them. That’s it. An exaggeration to be sure, but for the most part, I suspect, true.
September 9th, 2009 at 10:12 am
Well, I teach at a university, whereas the points here are obviously focused on primary and secondary teaching…but I can say that I think Matt may be onto something. I went to grad school pumped up to immerse myself in scholarship and change the world…fifteen years later, I’m rather weary of the fact that half of my students have absolutely no intellectual aspirations whatsoever (to them, college is just 13th grade+, a fun way-station on their path to a lucrative job). Now I’m seriously looking around at other options. Being paid real money to teach would change things somewhat.
In response to the comment above:
It isn’t true that the teacher doesn’t matter. But teaching well is a very rare and difficult thing in some disciplines, and even many smart and well-meaning people never seem to puzzle it out. Any evidence that tells you otherwise is suspect. Though it does seem to be true that pretty damn good teaching and pretty damn bad teaching seem to be roughly equivalent in the eyes of students.
September 9th, 2009 at 10:20 am
I haven’t read through all of the comments. However, I am a high school history teacher. Here is my take on “merit pay”.
I teach at a great public high school. It is one of the best in the state. Great kids, involved parents, a long tradition, just overall a really great place to teach. I am a pretty good teacher too (great AP scores, great evaluations, blah blah)
I started out at an awful middle school (2 years). I learned a lot but it was a tough 2 years before I lucked into a transfer. Last year I reflected on my middle school tenure and thought about going back for the challenge.
However, the district would have to pay me at least 1.5 times more and give me the power to build my middle school team of teachers. Other than those 2 things, there is no incentive for me to go to a struggling school for the same pay. None.
Also, the private schools in our area, very well respected, pay their teachers less than the public schools.
Teachers will put up with the low pay. We just want to be able to do our job and enjoy it.
September 9th, 2009 at 11:19 am
What I find utterly baffling about this commentary is that pretty much every single argument against quantitative measures of teacher quality—the metrics are all perverse, it will be used to advance the interests of those measuring, being rewarded stifles the sorts of creative performances we’re looking for—could be used, with at least as much potency, to argue against the existence of quantitative grades for students. Plenty of studies show that teacher-given-grades (as opposed to standardized tests, which have their own issues) can be influenced by such obviously screwy metrics as whether or not a male student has an uncommon first name, whether or not the student is attractive, whether or not the student resembles the teacher, etc., etc.
It seems like either we admit that imperfect metrics are a necessary evil for both teachers and students (and that the best way to deal with them is to have a LOT of imperfect metrics all taken into account, so at least their flaws will tend to cancel each other out), or we admit that we’re not up to the task of quantitative measurement in either case and abolish grades, period. But I don’t see how you can have it both ways.
September 9th, 2009 at 12:34 pm
Seniority is a very poor proxy. It is WELL-known and observed that many excellent teachers actually get worse over time, and start out really good folks who are so bad by the time they retire that their peers and administrators heave a sigh of relief once they’re out the door.
September 9th, 2009 at 12:37 pm
I agree with many of the criticisms made of Yglesias’ argument here (although not Sailer’s).
However, to try to be more constructive, let me lay out a way to harmonize the two viewpoints:
1. We need more good teachers, no matter what actually happens to the bad teachers, given the huge shortfalls we’re going to see when the baby boom generation of teachers retires.
2. We don’t actually know how to tell who is a good teacher.
3. In that case, the most rational course of action is NOT merit pay, but rather to increase base pay to the level of other professionals. By increasing the socioeconomic status of teachers, you will attract a horde of very smart and talented people. Now, not all of them will be good teachers, but many of them will be, and the teacher shortage will be dealt with.
September 9th, 2009 at 12:57 pm
[...] I wrote yesterday that it would make more sense to pay more money to more effective teachers, Steve LaBonne responded in comments that “Merit pay is a way for ‘reformers’ to [...]
September 9th, 2009 at 3:50 pm
We don’t actually know how to tell who is a good teacher.
Total hokum.
We may not have the perfect metric to measure teacher effectiveness, but the idea that we are completely incapable of determining who is a good teacher and who is a bad teacher is laughable. Your average third-grader can tell the good teachers from the bad.
September 10th, 2009 at 12:06 am
With all due respect, I’m astounded by the certainty that many posters have about the correctness of their position (”it’s the unions”, “it’s poverty,” “good teachers, bad teaches, it doesn’t really matter”, “it’s well known that…”,”your average 3rd grader….”, “most students do well, most schools are good”, etc.), particularly since I suspect that the “expertise” comes primarily from having been a student for quite a number of years (and perhaps being a parent of students), which, as we all know, makes one an expert on education.
Would the same people opine with equal certainty about matters medical, legal, or any other professional field? (No, I’m not saying “we educators know best.” I’m saying “save some room for some doubt and humility about a complex subject.”)
No, we DON’T all know what blend of ingredients makes for an outstanding, inspirational teacher, if for no other reason than what motivates and inspires ME to work hard to achieve more than I thought I was capable of may do diddly for you. And the teacher I thought was fabulous (or terrible) when I was a sophomore in high school I may quite well feel differently about when I’m 50 and have the perspective of many years of living behind me to better judge the impact that teacher had on my life.
Personally, I’m agnostic about “merit pay.” It will neither destroy nor save education. I DO know that most educators (including many of the fabulous ones who would clearly benefit from it) don’t like it, and I tend to be a bit wary of an approach that says “we outsiders know what’s best for you, dammit!” and then ram an approach down the reluctant throats of the “beneficiaries.”
One of the factors that makes evaluation of teachers difficult is that teaching is (most of the time) a solitary activity, unobserved (in its natural state) by other adults, which is different than the informal constant evaluations that occur when adults naturally work together in a work setting. Imagine a manager walking out of his/her office, clipboard in hand, who then sits in the outer office, a few feet away from the office support person. “I’m going to sit here for the next 45 minutes and watch you type, greet visitors, answer the phone, etc. Please try to ignore the fact that I’m here. And, I’m going to base my judgment of your job effectiveness solely on doing this 3-4 times over the next 6 months.” Who would consider that a meaningful basis on which to determine pay?
Let’s start with the agreement that we simply don’t (and probably can’t at this point) know what constitutes “great teaching,” and not try to answer that question by pulling answers out of the air or citing dueling studies. Then let’s allow experimentation rather than trying a global “cure” that may be worse than the disease, and not insist that we need to find a single “right answer” that will solve the myriad causes of declining student achievement.
September 12th, 2009 at 9:18 am
[...] Matthew Yglesias: Indeed, for all the controversy around differential pay schemes at some level I don’t think even the most old-school of teacher’s union leaders seriously dispute this logic. After all, it’s extremely common for collective bargaining agreements to offer enhanced salaries to teachers who have more educational credentials. The logic here, presumably, is that more educated teachers are more effective teachers and thus it makes sense to pay extra to retain them. The diplomas, in other words, are a proxy for quality. Similarly, veteran teachers get paid more than brand new teachers on the theory that a more experienced instructor is a better instructor. The principle that it makes sense to pay extra for quality isn’t seriously in dispute. The problem is that diplomas and time served turn out to be bad proxies for quality: “Recent research, however, suggests that such paper qualifications have little predictive power in identifying effective teachers.” [...]